1. Introduction
The goal of this research is to conduct a comparative analysis of two crowdsourcing-based remote work platforms. In the current online labor trade market, micro job markets and auction-based outsourcing markets play a significant role. However, there is a lack of user awareness regarding the differences between these platforms, and research on this topic is insufficient. This gap may be addressed using Bandura's self-efficacy theory and modern lexical analysis theory [1], enabling users to make more effective platform choices.
For this study, we chose to focus on Fiverr.com and Freelancer.com as representative crowdsourcing platforms. These platforms were selected due to their significant market presence and contrasting operational models. Fiverr [2], founded in 2010, is a leading micro-task platform with over 3.42 million active buyers as of 2021. Freelancer [3], established in 2009, is one of the largest freelancing and crowdsourcing marketplaces globally, boasting over 51 million registered users across 247 countries. While Fiverr operates on a gig-based model where sellers offer pre-packaged services, Freelancer uses a project-based model where clients post jobs and freelancers bid on them. This contrast allows us to examine how different platform structures might influence marketer self-efficacy and performance.
The methodology of this study includes surveys targeting active marketers on both platforms and an analysis of the lexical characteristics of press releases written by the marketers. Notably, the Remote Work Self-Efficacy (RWSE) survey is based on Bandura's theory of self-efficacy [1]. Additionally, lexical analysis is performed to assess the characteristics of the press releases written by these marketers. This methodology is expected to improve understanding of the self-efficacy and linguistic capabilities of marketers in a remote work environment.
The study includes 30 marketers active exclusively on one of two online work platforms, for a total of 60 users. The primary research task is to identify the differences in self-efficacy and lexical characteristics among marketers in remote work environments and to provide criteria to help users choose platforms that suit their needs. By analyzing data obtained directly from crowdsourcing marketers, this study aims to facilitate an academic understanding of self-efficacy and lexical elements in remote work environments and contribute to users' decision-making processes.
The remainder of this paper is structured as follows. Section 2 provides the theoretical background and related work. Section 3 describes our research design, including the methodology and data collection process. Section 4 presents our experimental results and analysis. Finally, Section 5 discusses our findings, their implications, concludes the paper, and suggests directions for future research.
2. Theoretical Background
2.1 Concept and Evolution of Crowdsourcing
Crowdsourcing, the practice of outsourcing tasks to a large group of people, has not only transformed remote work environments but also significantly impacted various other fields. This method enhances task decomposition and participation in software development, as Sharma et al. [4] noted, by enabling more comprehensive problem-solving approaches. Niu and Qin [5] observed that crowdsourcing improves the efficiency and quality of product design, though it also presents challenges in maintaining privacy and preventing cheating.
In the realm of data collection, Li et al. [6] identified crowdsourcing as a means of cost-effective and large-scale data acquisition. However, this approach necessitates the implementation of stringent quality control and privacy protection measures. As Whitla [7] noted, crowdsourcing in marketing is instrumental in product development, advertising, and research, with vast potential for future applications. Erickson et al. [8] emphasized that the ability to match people to their most appropriate jobs is critical to the success of crowdsourcing initiatives.
Hosseini et al. [9] proposed a reference model that categorizes crowdsourcing into four pillars, providing a structured framework for understanding its diverse applications. Erickson et al. [10] further explored the organizational uses of crowdsourcing by highlighting its productivity, innovation, knowledge capture, and marketing/branding benefits.
Integrating these insights into the study of remote work environments reveals the significance of crowdsourcing in modern marketing. The research aims to analyze the self-efficacy and lexical capabilities of marketers in remote work settings, considering the broader context provided by these studies. This comprehensive approach underscores the transformative impact of crowdsourcing across various sectors, especially in enhancing the effectiveness of remote work and marketing strategies.
2.2 Remote Work and Self-Efficacy
Remote work has revolutionized the dynamics of an effective workplace. However, while this effectiveness depends on a variety of factors, self-efficacy stands out as pivotal. Staples et al. [11] and Kondratowicz et al. [12] both underscore the significance of self-efficacy in successful remote work. Staples et al. [11] focused on remote work's impact on work effectiveness, productivity, and job satisfaction, while Kondratowicz et al. [12] cited its role as a mediator between remote work and job satisfaction.
Lange and Kayser [13] and Schoch [14] delved further into the relationship between self-efficacy and work-related stress. Lange and Kayser [13] discovered that self-efficacy not only reduces stress but also mediates health outcomes. Schoch [14] applied this observation to the appraisal and performance of information systems, highlighting the multifaceted influence of self-efficacy in remote work settings.
Practical tools for assessing and enhancing self-efficacy in remote work have been developed by researchers including Tramontano et al. [15] and Lathabhavan and Griffiths [16]. Tramontano et al. [15] created the e-Work Self-Efficacy Scale, a measurable approach to understanding self-efficacy in remote work contexts. Lathabhavan and Griffiths [16] identified technology, manager support, and peer support as crucial antecedents to self-efficacy, providing a broader understanding of the factors that foster self-efficacy in remote environments.
Lastly, the work of Grant et al. [17] brings attention to the importance of management trust and style in the effectiveness of remote work, emphasizing the need for a supportive and trust-based management approach to maximize the benefits of remote work.
These insights into the study of self-efficacy among crowdsourcing marketers in remote work environments clarify that self-efficacy is not only crucial for marketers' job performance and satisfaction, but also positively influences their stress levels and health outcomes. The measurement and analysis of self-efficacy performed in this study, in connection with lexical elements, are expected to provide a deeper understanding of how self-efficacy impacts the linguistic expressions and communication abilities of marketers, enhancing their overall effectiveness in remote work settings.
2.3 Importance of Lexical Analysis
Lexical analysis, a tool frequently used in online marketing, serves to extract valuable insights from vast amounts of textual data; this tool is a critical aspect of remote work environments that involve data analysis, particularly for crowdsourcing marketers whose tasks often involve analyzing large volumes of text for effective communication and strategy formulation.
Aggarwal et al. [18] and Mathieu [19] highlighted the potential of lexical analysis in determining brand positions and consumer perceptions. The insights obtained through such analyses are invaluable for marketers in remote settings, as they aid in customizing communication strategies to align with brand identity and consumer expectations.
Bharadwaj [20] demonstrated the effectiveness of lexical analysis in opinion mining and sentiment analysis, which are especially applicable in the context of online product reviews. For remote marketers, understanding customer sentiments through lexical cues is essential for developing targeted marketing campaigns and product improvements.
Christianto [21] emphasized the importance of lexical analysis in deriving marketing intelligence and identifying linguistic features of online advertising. This knowledge enables marketers to create more persuasive and resonant advertising to drive successful online marketing campaigns.
Additionally, Hynes and Janson [22] and Ma et al. [23] extended the application of lexical analysis to the effectiveness of Internet marketing and the extraction of product features from consumer reviews. Such capabilities may assist remote marketers to gain a better understanding of market trends and consumer needs, allowing the development of more effective product positioning and marketing strategies.
These studies underscore the significance of lexical analysis in several areas of online marketing, including recognizing brand positions and extracting consumer insights. From studies of crowdsourcing marketers in remote work environments, these insights into lexical analysis provide a robust framework for understanding how linguistic abilities and self-efficacy impact marketing effectiveness and communication strategies. The correlation between a marketer's lexical choices and their self-efficacy, as well as their ability to extract and utilize consumer insights from textual data, is expected to contribute significantly to enhancing the efficiency and effectiveness of remote marketing strategies.
2.4 Previous Studies on RWSE and Lexical Analysis in Digital Marketing
Several studies have explored the dynamics of RWSE and lexical characteristics in digital environments. Staples et al. [11] investigated the relationship between self-efficacy and remote work outcomes, finding that higher self-efficacy was associated with better performance and job satisfaction. Our study builds on this by examining how self-efficacy varies across different crowdsourcing platforms.
In the realm of lexical analysis, Aggarwal et al. [18] demonstrated that the lexical characteristics of online content can indicate brand positioning. While their focus was on consumer-generated content, our study extends this concept to marketer-generated content, exploring how platform choice might influence the sophistication of marketing language.
Wang et al. [24] examined self-efficacy levels among remote workers across various online platforms but found no significant differences. Our study's focus on specific marketing tasks on Fiverr and Freelance may reveal platform-specific effects not captured in more general studies.
Kondratowicz et al. [12] explored the role of self-efficacy as a mediator between remote work and job satisfaction during the COVID-19 pandemic. Their findings highlight the importance of self-efficacy in remote work contexts, which our study further investigates in the specific domain of crowdsourcing marketers.
By comparing two distinct crowdsourcing platforms and analyzing both self-efficacy and lexical characteristics, our study contributes to this body of literature by providing insights into how different remote work environments may influence marketer performance and communication strategies.
3. Research Design
3.1 Research Method and Procedure
The research method and procedure for this study are depicted in Fig. 1. The research involved marketers working on two different crowdsourcing-based remote work platforms, Fiverr.com and Freelancer.com. These platforms offer remote and autonomous work environments where marketers interact with clients through various services.
Research methods and procedures.
Initially, the researchers requested press releases from marketers active on one of these platforms, excluding those working on both platforms to avoid duplication. Following this, the RWSE survey was conducted with selected marketers, and the lexical complexity of their press releases was analyzed to compare the two platforms.
Figs. 2 and 3 illustrate the employment methods of marketers on each platform. In Fig. 3, one applicant was selected from several candidates to write a press release. The collected press releases served as data for lexical analysis.
We selected marketers who had been active on their respective platforms for at least 1 year and had completed a minimum of 50 projects. This selection criteria aimed to ensure participants had sufficient experience on their platforms.
In Fig. 2, 20 marketers were commissioned to write press releases for a set fee.
This study aims to explore the relationship between self-efficacy and lexical ability among marketers in remote work environments, contributing to enhanced communication strategies and efficiency.
The chosen crowdsourcing platforms, Fiverr.com and Freelancer.com, were selected for their distinct features. Both platforms are renowned in the crowdsourcing market and provide remote, autonomous work environments. However, their operational models differ significantly.
Fiverr.com is categorized as a micro-job market platform, primarily offering services to perform small tasks or projects. Marketers provide various services at fixed prices and clients purchase these services. This structure focuses on short-term tasks that may be completed quickly, offering a range of services at relatively low costs.
In contrast, Freelancer.com is considered to be an auction-based outsourcing market platform. On this platform, clients post projects and freelancers bid on them, proposing their services and prices. Clients then select the freelancer with the most suitable service and price. This auction-based model is better suited to providing customized services tailored to the complexity and requirements of projects, focusing on more long-term and specialized tasks.
The selection of Fiverr.com and Freelancer.com for this study is based on their differing remote work environments and task models, allowing for a comparative analysis of the influence of these factors on marketers' self-efficacy and lexical abilities. Understanding the distinctions between these two platforms enables a clearer identification of the factors affecting the behavior and performance of marketers in remote work environments.
3.2 Measurement
In this study, we employed the RWSE survey, adapted from Bandura's self-efficacy theory [1], to assess marketers' confidence in their ability to perform effectively in remote work environments. The RWSE survey consists of 15 items, each rated on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree). These items measure various aspects of remote work, including goal setting, time management, communication, and technical skills. We also analyzed the lexical features of press releases written by marketers using three key metrics.
& centerdot; Lexical density (LD): This measures the proportion of content words (nouns, verbs, adjectives, and adverbs) to the total number of words, indicating the information richness of the text.
& centerdot; Lexical sophistication 1 (LS1): This represents the proportion of sophisticated or advanced vocabulary used, reflecting the complexity of language use.
& centerdot; Type-token ratio (TTR): This measures lexical diversity by comparing the number of unique words to the total number of words, indicating the range of vocabulary used.
These evaluation indexes were chosen for their ability to provide comprehensive insights into both the self-perceived capabilities of remote workers and the objective quality of their written output, which are crucial factors in understanding performance in remote work environments.
It was specifically employed to assess and compare the levels of self-efficacy among marketers working on the micro-job market platform Fiverr.com and the auction-based outsourcing market platform Freelancer.com.
3.3 Press Release Data Analysis and RWSE Measurement
In this study, the virtual online service provider Ssolver.com was used as the marketing subject to analyze press releases written by crowdsourcing marketers. Ssolver.com operates as an online marketing-specialized micro-job market targeting English-speaking startup businesses and also provides educational services such as lectures and consultations to users interested in online marketing.
Table 1 presents the press releases about Ssolver.com written by marketers through Fiverr.com. This table includes press releases from 20 marketers across five countries. Marketers completed press release writing tasks in an average of 3.6 days. The average word count of the written press releases was 386.55 words, and the average score for the RWSE was 5.8.
This analysis aims to assess the effectiveness and efficiency of marketing content created by crowdsourcing marketers and to measure the self-efficacy of marketers in a remote work environment. By examining marketers’ content quality and RWSE scores, the study seeks to understand the impact of remote work conditions on the performance and self-perception of marketing professionals.
Table 2 displays the press releases for Ssolver.com written by a marketer through Freelancer.com. This table includes 20 press releases created by a single marketer residing in India, which were completed within three days. The average word count of these press releases was 389.70, and the average RWSE survey score was 5.0.
Press releases created through Fiverr.com
Press releases created through Freelancer.com
Through the analysis of this data, it was possible to compare the remote work self-efficacy and press release writing capabilities of marketers active on Fiverr.com and Freelancer.com. This comparison contributes to an enhanced understanding of communication strategies and efficiency in remote work environments.
4. Experimental Results and Analysis
4.1 Results of the Remote Work Self-Efficacy (RWSE) Survey
Table 3 displays the comparative results of the RWSE survey among marketers working on the Fiverr.com and Freelancer.com platforms. The analysis, which included 30 marketers from each platform (60 total marketers), revealed that out of 15 survey items, 12 showed no significant difference in score between the two platforms. However, in three specific areas: “setting goals aligned with client objectives,” “prioritizing tasks for effective time management,” and “sending documents via email,” marketers on Freelancer.com scored higher on average with statistically significant differences.
These results demonstrate statistically significant differences in self-efficacy levels among marketers across the two platforms, specifically in three key areas. The higher average scores for Freelancer.com marketers in “setting goals aligned with client objectives” (t = -2.19, p & lt; 0.05), “prioritizing tasks for effective time management” (t = -2.22, p & lt; 0.05), and “sending documents via email” (t = -2.17, p & lt; 0.05) suggest that these marketers may possess clearer goal-setting abilities, more effective time management, and smoother document transmission skills. The moderate effect sizes (Cohen's d & approx; 0.5) for these differences indicate a practically meaningful distinction between the two groups. This variance in self-efficacy levels could be attributed to differences in the platforms' work environments, task structures, or user demographics, though further research would be needed to establish causal relationships.
This variation likely reflects the influence of each platform's work environment, the way marketers approach their work, and marketers’ perception of their abilities.
Thus, this analysis provides important guidelines for users in selecting a remote work platform that aligns with their work requirements and style. It may be beneficial to consider these differences in self-efficacy when determining work execution strategies and communication methods in remote work environments. This may offer significant benefits in enhancing the effectiveness and satisfaction of marketers working remotely, potentially influencing their job performance and overall contribution to online marketing projects.
4.2 Lexical Feature Analysis Results
In this study, a comparative analysis was conducted of the lexical characteristics of press releases written by the Fiverr and Freelancer groups. Table 4 shows the results based on LD, LS1, and TTR metrics.
The analysis revealed that the Fiverr group exhibited a higher LD compared to the Freelancer group, suggesting that their press releases contained more densely packed information. The Fiverr group press releases also showed higher LS1 than those of the Freelancer group, indicating a more professional and sophisticated use of vocabulary. Additionally, the Fiverr group showed higher lexical diversity compared to the Freelancer group.
RWSE survey analysis results between platforms (n = 60)
Inter-platform press release vocabulary complexity analysis results (n=40)
Overall, the press releases from the Fiverr group were more complex and diverse in terms of vocabulary than those from the Freelancer group. These results suggest that marketers in the Fiverr group use more professional vocabulary choices and a variety of expressions in their press releases. These lexical characteristics provide significant insights into how the lexical abilities and effective communication strategies of marketers may differ in remote work environments, illustrating the importance of tailored language in professional communication.
5. Discussion and Conclusion
This study aimed to compare and analyze the self-efficacy and lexical characteristics of marketers active on two distinct crowdsourcing-based remote work platforms, Fiverr.com and Freelancer.com. The research was intended to explain the various factors influencing marketers' behaviors and performances in remote work environments. The self-efficacy analysis revealed that marketers on Freelancer.com scored higher in certain self-efficacy items compared to those on Fiverr.com, indicating greater confidence in goal setting, time management, and effective communication. In the lexical characteristics analysis, press releases written on Fiverr.com showed higher levels of LD, sophistication, and diversity compared to those on Freelancer.com, suggesting that Fiverr.com marketers tend to use more professional and varied vocabulary in communication.
This study provides insights into the differences in marketers' self-efficacy and lexical abilities in remote work environments and how these differences may impact marketing performance. Both self-efficacy and lexical ability play important roles in successful work performance and effective communication strategies in remote work settings.
These findings offer useful guidelines for users to choose a remote work platform that matches their work requirements and style, and indicate the importance of remote work marketers' self-efficacy and lexical abilities when determining work execution strategies and communication methods. This research may serve as foundational data for improvements in future remote work strategies and marketing communication.
However, this study has limitations. It focuses on two major platforms (Fiverr.com and Freelancer.com), excluding other important ones like Upwork, which may limit generalizability. The relatively small sample size, reliance on self-reported data, and potential influence of subjective evaluations and cultural or regional factors constrain our findings. Our study also primarily focused on short-term outcomes and did not extensively explore psychological and social factors or platform-specific features.
Our study provides insights into the differences between Fiverr and Freelancer, but these findings may not be fully generalizable to all crowdsourcing environments. Each platform has unique features and user bases that could influence marketer performance and self-efficacy. Our results should be interpreted as a comparison between two specific models of crowdsourcing platforms rather than a comprehensive analysis of all crowdsourcing environments.
Future research should address these limitations by expanding to other platforms, increasing sample sizes, incorporating objective performance metrics, and using longitudinal or mixed-methods approaches. Studies should also consider cultural and regional factors, long-term success factors, and psychological and social aspects of remote work. This would provide a more comprehensive understanding of how different platform designs and diverse backgrounds influence marketer self-efficacy and performance across the gig economy landscape.